JOURNAL OF RESEARCH of the National Bureau of Standards

Volume 82, No.2, September-October 1977

Fundamentals of Building Heat Transfer

Tamami Kusuda

Institute for Applied Technology, National Bureau of Standards, Washington, D.C. 20234

(July S, 1977)

Basic proble ms and unique features of building heat tran sfer a re described in rela tion to th e heating and
cooling load calcu lation, whi c h is a sta rting point for building e ne rgy cons umption analysis a nd eq uipme nt
s izing. De tailed discussion is given of the rela ti o nship between heat loss (heat ga in) a nd heating load (cooling
load ). Al so outlined is a discussion of the multi-space heat tra nsfe r proble m in whic h th e a ir a nd heat exc ha nge
equation s among adjacent spaces in a building are solved simultaneous ly with th e radi a nt heat exchange
equ a ti o ns for the s urfaces of each room.

Key words: Air-l ea kage; d ynami c heat transfe r; e nergy analysis; hea t a nd cooling loads; heat loss a nd heat ga in ;
multi - roo m probl e ms.

Nomenclature

A Matrix element as defin ed in th e tex t QS Ne t long-wave radiant heat to an ex te rior su r-

B Ma trix element as defin ed in th e tex t fa ce
Cp Specific heat of air S int e rior surface area
DB Outdoor dry-bulb te mperature TA or TA Space a ir te mperature
F Exterior s Ulface convective hea t tran sfer coef- Tf Inte rior s urface te mpe ra ture
fi cie nt TO or To Exterior surface te mperature
G Mass air flow be tween adjacent spaces TS Supply air te mpe rature to a space
H Inte rior s Ulface co nvective heat tra nsfe r coeffi- X n, Yn, Zn (n = 1) . .. N thermal response /actors
cie nt a solar absorptance 0/ exterior su1ace
Inte rior s urface radiative heat transfer coeffi- T solar transmittance 0/ window
cie nt Subscripts
Solar rad iat ion indicent upon an ex te rior s ur- t time
face Superscripts
QG Strength of hea t so urce/sink ~, } de notes ith andjth spaces
r Radi a nt heat in cide nt upon each interior s ur- k, I de notes kth and Ith s urfaces
face

1. Introduction • the rmal storage in ex te rior masses of buildings.

The purpose of this paper is to summarize state-of-the-art (b) Interior environmental analyses -
information on various aspects of building heat transfer, and • radiant heat exc hange among interior surfaces
to di sc uss two selected subjects in detail; namely, the multi- and heat sink/sources,
space heat transfer problem and the relationship between • convective heat e xchange between room air a nd
heat Joss and heating load. room interior sUifaces,
Building heat transfer calculations are pe rformed for • room air convection . . . inter- and intra-room
differe nt applications such as: convective motion,
• convective and radiative heat tran sfer of inte rn al
(a) H eat loss a nd heat gain through exterior e nvelope- heat sources such as heaters, cool e rs and occu-
• condu ction through exterior envelope, pants,
• condu ction heat transfer through baseme nt walls • thermal storage in interior masses.
and slab-on-grade floor (. .. to se mi-infinite
region) , (c) Material or building element-related proble ms-
• s hort wavele ngth (or solar heat) transmission, • cold-bridge effect,
absorption, and refl ection for fen estration, • convection within porous ins ul ation,
• air leakage through exterior e nvelopes as well as • moisture conde nsation du e to s imulta neou s flow
th e interior partition walls, ceilings and floors , of air, moisture , and heat.

97
Basic to all of these applications is the fact that heat transfer (a) Heat transfer at the exterior surface considering
processes for buildings are usually ill defined, time depend- both wind and surface roughness characteristics,
ent, multi-dimensional, and in many cases non-linear. Thus
all the solutions available today for anyone of these (b) Convection in and through the cavity walls,
applications are based upon numerous simplifying assump-
tions. In many cases refined and advanced solutions may be
available, but they tend to be overly complex and have very (c) Convection between the window glass panes,
little practical value. In addition, until recently, at least in
the United States, most applications of advanced heat-trans- (d) Inter- and intra-space air motion due to temperature
fer analysis have traditionally been explored by the aerospace and pressure gradient,
and nuclear power industries. For this reason, many of the
challenging problems in building heat transfer have been (e) Convection heat transfer due to air leakage through
left unsolved.
exterior walls.

2. Conduction Heat Transfer

(f) Convection heat transfer within the porous insulating
Conduction heat transfer problems relevant to buildings structure.
include:
The surface heat transfer coefficients are affected by the
(a) Exterior wall conduction- nature of the air boundary layer which is strongly influenced
• transient heat transfer responding to climatic by the surface geometry, temperature gradient, and the flow
effects, such as temperature fluctuation, solar outside the boundary layer region. Unfortunately most of the
radiation, wind and precipitation; thermal stor- textbook solutions and analytical expression available are
age . . . damping and lag effect; and cold-bridge based on very simplistic boundary conditions. They should
effect (two-dimensional and non-linear heat flow be considered as only approximate solutions to the real
path). building problems. The air flow around the building, along
the internal surfaces, and in the wall cavities is very
(b) Interior mass conduction- complex, erratic and not conducive to exact analytical
• heat storage in partition walls, floor/ceiling sand- solutions. Moreover, the actual building surface geometry is
wich. seldom well defined.
Many inexperienced building heat transfer analysts have
(c) Conversion from heat gain/loss to cooling and heating mistaken notions that the exterior surface or interior surface
load. heat transfer coefficients as found in the heat transfer
textbook or in the ASHRAE Handbook of Fundamentals [5]
(d) Ground heat loss from slab-on-grade floor and base- are absolutely accurate. This notion was conveyed to the
ment walls. author recently in one of the thermographic heat-flux mea-
surement meetings, at which an attempt was made to convert
Most of the building heat conduction problems are multi- the surface temperature measured by a sophisticated infrared
dimensional and transient, requiring numerical (either finite scanning apparatus to a heat flux value by applying the
difference and/or finite element analyses) or analog com- ASHRAE Handbook value [5] for the surface heat transfer
puter-simulation calculation. For one-dimensional multi- coefficient. In reality, the surface heat transfer coefficient
layer problems, numerous analytical solutions are available, could differ from the published value by as much as 1000/0,
for both steady-state and transient conditions. These solu- depending upon the local wind gust and the irregularity of
tions are frequently used for heat gain and loss calculation the surface geometry. The fact is that, except those reported
for exterior envelopes and heat storage in interior structures. by Ito and Kimura [6], the exterior surface heat transfer
Most notable contributions in recent years for the multi- coefficients for actual buildings have never been studied
layer problems are the thermal-time-constant concept of experimentally. The same is true for the interior surface
Raychaudhuri [IV the admittance solution of Laudon [2], heat transfer coefficients, except for the original work of
the frequency response solution of Muncey [3] and the Wilkes and Peterson [7].
response factor solution of Mitalas and Arseneault [4]. These A number of studies have been reported in recent years in
solutions permit the accurate evaluation of heat conduction which numerical solutions of the Navier Stokes fluid dynam-
through building walls, roofs and floors so long as the heat ics equation and the e nergy equation were obtained for two-
flow is normal to the surface. Missing from these methods dimensional air flow within a confined space for the laminar-
and awaiting development are similar solutions for multi- flow regime characteristics of room convection, and into the
dimensional problems related to building corners, floor slab- turbulent-flow regime (the Grashoff number of 1012 or larger).
on-grade, and basement walls. In addition, steady turbulent-flow solutions to the modified
Navier-Stokes equations were obtained by Nielsen [9]. The
3. Convection Heat Transfer computer time and computer memory size ~equired to deal
with realistic three-dimensional flow problems in a room are
Convective heat transfer problems relevant to buildings formidable. Bankvall [10] and Tien [11] are studying flow
include: equations within porous media, and Tien is extending the
solution to partially filled insulation cavities, as well as to
the permeable boundary surface problems, to simulate air
I Figures in brackets indicate the literature references on page leakage in insulated walls.

98
 4. Radiant Heat Exchange age ope rating conditions, the heatin g and cooling equipme nt
designed or selected is usually ove rsized a nd operates at
Radia tion heat transfer is very important in building part-load condition s that result in poor effi c ie ncy .
appli cation in the following a reas: On the other ha nd , the European a nd Ja panese engineers
have been predi cting indoor te mpe rature as a fun ction of the
(a) Shor-t-wavelength rad ia ti on: ever- changing outdoor climate conditions. Th e ir interest has
• solar heat absorption on opaq ue exterior sur- been prima ril y to dete rmine th e need for air-cond iti on ing by
faces, estimating the numbe r of hours in th e sum mer when the
• solar heat trans mi ssion through transparent sur- indoor tempera ture could exceed the comfort req uirement.
fa ces, Since most No rth Ameri can homes and co mmercial build-
• solar heat absorption and reflecti on by interior ings are already a ir-conditioned and centrall y heated , major
building sUlfaces, emphasis in the Uni ted States is to be able to predict e nergy
• absorption and reflection of solar heat by win- consumption by thi s heating a nd cooling equi pme nt. It has
dow glass. been well accepted in th e United States th at th e most
acc urate way of es timatin g th e annual ene rgy consumption
(b) Long-wavelength radiation: of a given building is to simulate th e heat transfe r perform-
• heat e mission by th e ex te rior surfaces to the ance of th e building and the performan ce of its heating,
sky , ve ntilating a nd air conditioning syste ms on an hourl y bas is
• heat excha nge a mong interior surfaces, throughout th e year, thus requi rin g 8760 calculations. Dur-
• heat exchange between interior surfaces and in g the past d ecade, a numbe r of sophi sti cated computer
occupan ts, progra ms have been developed to do th e hourly s imulation
• heat exchange between the li ghting fixture and [1 3]. The Ameri can Society for Heating, Refri geratin g and
interior surfaces. Air-Conditioning E ngineers has, for example, developed
recommended procedures [14, 15] to s imulate the building
heat transfer, a nd system a nd equipm ent performances .
A diffi cult portion of this hourl y s imulation approach is
the coupling betwee n th e building heating and cooling
Whil e extens ive work has been reported in the area of solar requireme nt and the available capabiliti es of th e building
energy exchange with exterior surfaces, bas ic irradiation heating and cooling syste ms- es pec iall y for th e building
data are still insufficien t with respect to the diffuse s ky with many zones, each requ iring different heating and
radi a tion component , particularly for ve rtical surfaces a nd cooling with respec t to time-dependent use schedule a nd
for cloudy sky conditions. occupancy patterns . The coupling calcula tion becomes es-
Perhaps the most diffi cult and tedious problem in dealing pecially diffi cult wh e n the heatin g and cooling capacity
with solar heat exchange a nalysis for building applications provided by the heat di stribution syste m and by th e building
is the analys is of direct or bea m radiation that is trans mitted central plant cannot match with the space heatin g and
throu gh fenestration , abso rbed , reflected and reemitted by cooling requireme nts.
the interior surfaces. Because of the c omplex geometry of Thus when a mi smatc h between th e calculated require-
the time dependent shade and sunlight patterns, exac t ment and th e available system capacity occurs, the space
solutions to s imula te realisti c solar heat exchange in a room te mperature drifts from th e set-point for whi ch th e original
are virtually impossible. It is usuall y assumed that the load calculation was performed. The a mount of this drift is
interi or surface is gra y (non-spectral) and diffuse (no specular depende nt upon the the rmal storage characteris ti cs of the
refl ec tion), and th e solar heat is diffused once it enters space under consideration, the operati ng characteris tics of
through the window and interior shading devices . the heat distribution syste m, and the part-load or the overload
The long-wavele ngth rad iant heat e mi ssion to the sky charac teristics of the central heating and cooling plant. To
from the building exterior surface has not been well explored the best of this author's knowledge, no existing energy
except for the clear s ky condition , although some work is analysis computer program can handl e thi s problem in a n
going on at CSTB in France [12] . exact manner.
Simultaneous calculations of multi-space heating a nd
5. Heating and Cooling Loads Calculations cooling requirements, together with the heating and cooling
capability of central HV AC syste ms, under dy nami c condi-
Because of the increased urgency for energy conservation tions, (where the climate as well as operating parameters are
in th e design of buildings, and because the essenti al part of constantly changing) are extremely diffi cult. Thus a common
the energy conserv ation design is th e accurate dete rmination Ameri can practice is to calculate the space heating and
of building heating and cooling require ments, a great empha- cooling requirement at a give n set of required space temper-
sis has been placed upon load calc ulations in recent years. ature condition s. The coupling between the load and the
It might be mentioned tha t at leas t in the Un ited States system capacity is evaluated by the use of the weighting
past prac ti ces for heatin g and cooling load calculations we re factor concept. The weighting facto rs are trans fer fun ctions
based upon steady-sta te heat transfe r pel{ormed at design generated by Mitalas [4] for three typical rooms representing
conditions that represent ex tremely cold and hot days . This li ght , medium and heavy construction, by solving d etailed
is because the heatin g and cooling loads in the United room surface heat balance equations .
States we re calculated solely for th e purpose of selec ting S.l. Heat Balance Equation at Exterior Surfaces
and sizing the heatin g and coolin g equipment to provide
comfort for the ex tremes of climatic conditions . Sin ce these The building exterior surfaces receive sol ar radiation,
extreme design conditions rarely represent day-by-day aver- exchange long-wavelength radiation with the surroundings

99
including th e sky, exchange heat with th e outside air by th e and
convection process , and conduct heat into th e solid s tmc ture.
Equ ati on (1) represents a complete heat balance on the AI(i,k) = H(i,k) .
exterior surface with th e response factors being used for the
heat conducti on term [4].
Equa tion s s imilar to (3) should be prepared fo r all th e
interior sUlfaces in a give n space (room). On th e oth er hand ,
qii,k) = I y<~,k). mic.}! th e ove rall heat balan ce equ ati on of th e space air should
n=o
include the convec tion heat exchange with th e surfaces,
heat sources, air leaked into th e space, and heat give n off
I Z~,k). TO~ic.~) (1) and suppli ed by th e heat sources/si nks (incl udin g s pace
conditioning devices), as follow s :
n=o

N i
I Ji(i ,k)(Tlii.k) - TAii»)S(i,k )
k=1
5.2. Heat Balance Equation at Interior Surfaces
Ni
(Inter-Space Heat Exchange)
+ I GC~ , k) (A1/i,k) - TAi i ») + QG(i) =0 (4)
k=1
Equation (2) shows all th e co mpone nts involved in th e
h eat balance of an interi or surface; namely, th e long-wave-
le ngth radiant heat excha nge with th e rest of th e surfaces, By letting
c onvective heat exchange with th e air, in cident solar radi a-
tion (through the windows), long-wavelength radi a tion from
li ghting a nd equipment , and heat stored into th e solid
ma terial , which is expressed in response-fac tor terms . Mi = Ni +1

qii,k) = I X~ ,k)T/~ic.}!
Nt Ni
n=O
Afi,Mi) = - I /f i,klS(i, k ) - I Gi,k . Cp
k=1 k=1
- I }'l~ ,k)TOiic.}!
n=O
(2)

Ni
+ I If;i,k)(Tf/,j) - T~ ,k») + rii,k) The space heat balance equ ati on (4) th en becomes :
j=1
N i
By rearranging the terms, eq (2) becomes I A~,Mi)TIi,k + A/(i,Mi) . TAi i )
k=1
N i Ni
I A~i,k)mi ,j) + Af i ,k)TA~i) + I AO<~,M;)'A1/i,Kl = B(i,M;l (5)
;=1 k=1

k = 1, Ni (3) A set of (Ni + 1) simultaneous equati ons compnslllg Ni

equations of type (3) and equ ation s (5) must be solved
where Ni total number of h eat transfer surfaces in th e ith
simulta neously for a give n space to yield the interior sUiface
roo m
temperature and space air temperature .

B (i,k ) = I X~,k)mic.~)
5.3. Inter-Space (Inter-Room) Heat Exchange
n=1
Whe n th e space un der considera ti on in th e previo us
+ I y<~ , k)TO~ic.~) + rii ,k) section is adj acen t to oth er spaces, all of th e inner sUiface
n=1 te mperatures and air temperatures fo r all of th e adj ace nt
Ni
spaces are being affected by one another. Figure 1 illustrates
A~i ,k) = X~ , k) + /fi,k) + I If;i ,k) the heat transfer process through a boundary wall between
j=1 th e ith s pace a nd jth space, indicating condu cti ve and
convec ti ve heat exchange between th e tw o spaces . A simpli-
fi ed and yet relatively typical four-space p ro ble m, as de-
picted in fi gure 2 , is used herein to illustrate th e complex ity
of the multi-room problem.

100
 i-th space j':'th space
k-th surface I - th surface

N
Air Temp Air Temp j

Radiati ng
Sur face s (. 1)
TIl,

Radiant heat
/ t
Ni transfer coeff i cient
Radiatin~
Surfaces
/
f1 ( i , 1)
1 TI(; , 2)

H( '; ,1 ) _ _ _ _ _ _ _ t

H(i,k)
4

H
N.

Radian:~~ TI(;'N)
r (; ,1 )heat
t t 1

Sur face Temper at ur e

Surface Temperatur e
To(i ,k )
t

response Surfa c e Convection

Surface Convection
factors heat transfer coeffi c ient
he .t trans f er coeff i c i ent
He; , 1)
X(i,k): Z(j ,1)
H(i,k) o
n n
i
y(i,k): y(j ,1)
n n

Z(i , k): xC; ,1)

n n

FIG URE 1. Heat transfer oj a partition wall between th ith andjth space.

In ord er to simplify the mathematical ma nipulati on, figure pand ed a compre hensive computer program to calcula te air
3 is provided to show th e interaction of four spaces by flow between building spaces as affected by the ex terna l
th erma l couplin g through adjacent inte ri or surfaces . Figure wind pressure as well as by the therm al stac k effect. Th e
4 shows the matri x form of the complete hea t balance program was primarily developed for the stud y of s moke
e quation to be solved for thi s four- space probl em using migration pattern analysis under a fire researc h program. It
matrix nota ti ons prov ided in eq uations (2) and (3). In thi s does not incorporate the th ermal storage cha racte ri sti cs of
matri x presentation a ir lea ka\ie between th e a dj ace nt spaces the building internal thermal mass ; however, th ere is a real
i s presented as a consta nt G(' ). Air leakage is, however, in need for an effi c ient co mputation procedure by which th e
reality a function of the te mperature difference and pressure multi-space air a nd heat transfer process can be a nal yzed in
difference between the space du e to wind effect. Tamura a comprehe nsive manner.
and Sanders [16] developed and later Fothergill [17] ex- The above discussions indicate that a co mprehe nsive

101
 SPACE 1 SPACE 2 SPACE 3 SPACE.4 ORIENTATION

NORTH
,.
SPACE 3 SPACE" 01
02
I
I
I
. ~1
\ ,
0 1
02
SOUTH

,. I-7"""""-.,.,-.,-,--,-,-:---,----,--'"'--:--:--:-,------,.-,--:....,.--,---,-t ;. SOUTH \
••1 '. _, ' ,_ " " ' .~' / .~ ..... : _ • \. . ". i .' " .'.' .
I \
03 I 02 02 \ 03 WEST
I \

.
I \
I \

••
SPACE 2 SPACE 1
I \
.: I

L....,-:-:-.,..,.....,---,--~-:-:--::-:-"""-~--::---:--:-:-~ .'
03 03 • NORTH
.
, -
,: ... ,. " ' .". " . '. - -' : ,~ ..-

05 04 0 4 05 EAST
NORTH 06 Os 05 06
"\ ',' '.', -.', .'

WEST
..:. . . ......-..6,
...... ,,......
06 0 7 ROOF
/CEILINGI

Da 07 • ~........... FLOOR

o OPAQUE WALL .... EXTE.1011

;i"
~~ '. :- .", .: • OrAQUE WALL .... INTE.IOR

o aANSPARENT WALL IWINDOW)

SOUTH
l!.. OI'EN PASSAGE
FIGURE 2. PLan of a sampLe 4-space modeL.
FIG URE 3. Interior-exterior waLL relationship of a 4-space model as depicted
in figure 2.

analysis nm-linear of multi-room heat transfer problems Difficulty arises, however , in simulating th e regulation or
requires a large-size matrix manipulation, whi ch requires a control of GS and TC in the computational process . The heat
large and fast computer. The complexity of th e analysis is exchanges among th e space th ermostat, s pace a ir and room
basically du e to the inclusion of th e interior surface radiation surfaces enter th e heat balance eq ua tions, thereby increasing
heat exchange terms. Considerable simplification is possible, the complexity of the proble m. In addition , du e to th e time
if it can be assumed that all the room interior surface response charac teristic of the thermostat , th e s imulation of
te mperatures are the same as th e room air temperatures. A th e control system requires minute-by-minute calculations,
brief discussion of this subject is provided in th e appendix. rather than the hour-by-hour calculations.

5.4. Coupling With the Central System Capacity

Heat source/ sink term QG indicated in eq (4) co uld 6. Summary

represent th e heating and cooling capacity of the HV AC
supply air syste m. If th e supply air quantity and its temper- Basic problems and diffi c ulties inheren t in th e comprehen-
ature are expressed as CSt and TC t respectively, th e term s ive analysis of building heat transfer are du e to many
QC takes on the value CSt ' Cp ' ( TC t - TAt) . paramete rs which are ill defined and some of which are time
In this manner the s pace air temperature can be deter- dependent and multi-dimensional.
mined as a balance between th e space heating and cooling One of the crucial areas {or bu ilding heat transfer analysis
requirement and th e h eating and cooling capacity of th e is the hour-by-hour simulation of the balancing process
HV AC supply air system. Th e values of CS and TC may between the heating and cooling requ irements of man y
vary, however, according to th e mode and type of the s paces and the central heatin g and cooling capacity of th e
building air distribution system. For example, a constant air building. Analytical solutions available for a ny of these
volume system would provide GS constant and regulate TC, problems are based upon simplish c assumptions.
whereas a variable volume system would regulate CS with Detailed di scussions of radiant heat exchange effects for
TC held constant. building heat loss calculations are included in the appendix.

102
/
A~l,l) A~l.l) A~l, \' A!I,I I A~1 ,I} All,l) A~] ,I) ",!1.1 1 AP·1l 0 T,O,I) aO'I~
A~l.2) A~I,2) A~l , 2: ,,11 .2> A!I . 2),,!1. 2 ) A~ 1 . 2) ~! ,2),,11 , 2) 0 TIO,2) a(1.2)

T10.) a O ,)

TI(I,4) eO •4)

TI(1·H 3°' S)

T,(1 , 6) e(1,6)

0.4,6' · 7) 0 T10,7) a (l , 7)

TI O , 8) a O ,8)

TA CI ) a O ,9)

T1(2,1) a C2 ,1)

TI(2 , 2) a(2 , 2)

a(2,})
o A:2.) .... ~2.1)A~2.) ...!2,]1 ... ~2 . 1) A!2 , ) .... !2·)),d 2 . l) 0
TI(2 , )

a(2 , 4)
T1(2,4)
o . . :2 . 4) ,,~2, 4) ,,~2 ,4) ,,!2 ,4) ... ~ 2 ,4) .... !2 ,4) ... ~2, 4) AP ,4) 0

TI (2.S) 1(2, ~ )

T1 fl •61 a O ,6)

X T1(2.7) a(2,7)
r A;2 . 7) ... ~2 , 7)1, ~2 , · ) .<2 . 7)r.~2 , 7),,!2·7)A~2 . 7) .... f2.71 C
TA(2) ,(2 , 8)
o ... 6i,8)A:2.8)A~2.8)A~2,8) ...!2,8) ... ; 2 .iI)A!2,8) ... :2,8) ... p.8)
o ... ~l.l) 0 TI(},I) aO,1)

T10,2) II Cl , 2)

TI O , ]) ,(l,ll
o A:l'll ... ~1.1)A~)·Jl!)·}),,!1.1) ... !3.)IA:)·n . . p.n
TID , ') eO,,)
o ,,:3,4) ... ~1.4)A~}'\1!) . 4) ... ;) . 4) ... !).4)A:).4)AP .4 ) 0

TI P , S) a O ' S)

TIO . 6 ) e(],6)

Tl {),7l aO,l)

TAD) 8 0 ,8)

o ",:4 ,I) A~4 ,I) ... i4 ,I) ...!4 ,n ...!4 ,1) ...!4 ,I) Ai4, 1~!4.I) . . f4,1) T1(4,l) a(4,1)

... :4 , 2) A~4,2) Aji. ,2) <4,2) A~4 . 21,,!4 , 2) A~4,21!,,2) ... f4,2) T1(4,2) 11(4 . 2)

... :4 , 3)~4 '1) ... j4 . 1)A!4'])A!4'])1t.!4 , )) ... ~' . ll!4 . )) . . : 4 .]) T1(4.]) a (4 ,J )

o ,,:4 , 4) ... ~4 , 4) Aj4 , 4) ,,!4 , 4) ... !4 . 4) A!4 , 4) A~4 ,41!4 ,') "f4, ,) T1(4 ,4 ) a(4 . 4)

o "~, , S) ,,~4 , ~) ,,~~ , ~) ,,!4, 5) ,,~, , ~) A!' .~) A~4 , 5l!4 ,5) Ai4 , 5) T1(4,5) 8(4. 5 )

TI(4,f) 8(4,~)

n A~4 'IlAr' 7l A~" 71A!I, 71 A~4, 7)A!I., 7\j4, 7 l~/" 7 \ AV , ;) T,(4,7) 1(4,7)

o A:4 , 8) A~4 , 8) Aj4, 8) A!4 ,8) A~~ ' 8) A!4 ,8) A!4 ,8l~4 , 8) Af4 ,8) TI(4,8) a(4,3)

o ~6~,9 )· ,,:4 . 9) A~4 , 9) Aj 4 ,9 )A!4,9 ) A!4 ,9) A!4,9J A~4,9l! 4,9) Af4 . 9) TA(4) 1(4,9)

FIGURE 4. Heat baLance equation of a 4-space modeL.

103
 7. References This simple relationship is incompatible with reality,
where th e space su rface temperature is seldom equal to th e
[1] Raychaudhuri, B. c., Transient the rmal response of enclosures; the s pace air temperature .
integrated the rmal time constant, Interna tional J. H eat Mass Trans- To examin e th e difference in results be twee n the com mon
fer 8 , 1439 (1965) .
[2] La udo n , A. C. , S umm ertime T e mpe ratures in Buildings, Building assumption and th e exact solution, a s imple room of one
R esearc h Studi es, 1968. Building Research E stabli s hm ent , Build- si ngle-glazed window was s tudi ed. In thi s room the walls are
ing Research Station , Carston , Watford, WD2 7.JR. assumed massless and have th e same overall and s urface
[3] Muncey , R. W. , The thermal response of a building to s udde n hea t transfer coeffici e nts and the same interior surface
c hanges of te mperatures or heat Oow, Austra li an J. Sc i. 14, 123,
1963. temperatures , and are ex posed to the same outdoor air
[4] Mitalas, G . P., and Arsenault, J. C., Fortran IV program to calculate temperature as shown in figure A-I. Ignoring th e solar heat
Z-transfer fun c ti ons for the ca lc ulation of transie nt heat transfe r absorbed by th e glass, th e heat e xcha nge equation for the
through wall s a nd roofs , use of comput e rs for e nvironmenta l inner surface of the window for the model will be:
e ngineering related to buildings, NBS-BSS 39, 1971 , pp. 663-668.
[5] American Society for Heat ing, Re frige rating and Air Conditioning
Engineers, 1977.
[6] lt o, N., a nd Kimura, K. , Field Experime nt Stud y on the Convec tive
H eat Transfer Coeffic ie nt of Ext e rior Surfaces of a Building,
ASH RAE Tra ns . 78, 1 (1970).
[7] Wilkes, C . B., a nd Petersen, C. M. F. , Rad iation a nd convec tion
from s u,faces in various pos iti ons, ASHVE Tran s. 44, 1, (1938), where H G co nvec tiv e heat transfe r coeffi cient over the
pp. 513 . interior s urface of th e window.
[8] Fromm , J. E., A nume ri ca l met hod for computing the non-linear, radiative heat transfer coefficient of th e win-
time dependent buoyan t c irculation of air in rooms, use of comput ers
for en vironm ental e ngineeri ng re la ted to buildings , N BS-BSS 39, dow to th e interior surface of th e opaque wall.
(197 1) pp. 451- 467. thermal co ndu c tan ce be twee n the inne r s ur-
[9] Nielse n, P. V., Flow in Air Conditioned Rooms, Ph. D. Thesi s , face of window and outside ambient
T echni ca l Unive rsity of Denmark , Nordbe rg, 1974. interior surface te mperature of th e window
rIO] Bankva ll , C. C. , Na tura l Convec ti ve H eat Transfer in Ins ulated
Structures, Lund Institut e of Technology , (S we de n) Re port 38, interior s Ul{ace temperature of dl th e walls
1972. room air te mperature
[11] Tien , C. L. , a nd Burns, P. J. , Convection in a Ve rti cal Slot Filled outside ambient te mperature.
with Poroll s Insulation , Interna tional Heat a nd Mass Trans fer ,
Dubrovnik, Yugos lavi a, 1977.
[12] Bertolo, L., Instrumentati on for Measuring H e at Transfe r to th e Sky , To
an unpubli shed CSTB doc ume nt , CSTB , France, c irca 1975 .
[13] Crall, c., Bibliography o n Available Computer Programs in th e
Genera l Area of H eating, R efrige ra ting, Air Conditioning a nd
Ventilating, ASH RA E Researc h Project Report CPR-153, October
1975 .
[14] Task Croup on Energy Req uireme nts , Procedure to De te rmine H ea ting
and Cooling Loads for Comput e rize d Energy Calc ul a tion s for Build-
ings, special ASH RAE Bulletin, 1974.
[1 5] Stoec ke r, W. c., Procedure to S imula te HVA C Sys tem a nd Equipment
for Computerized Energy Calculations, Specia l ASH RAE Bulle tin ,
1974.
[16] Tamura , C., and Sanders, D. , A Fortran IV Progra m to Simulate Air
Movement in Multi-St ory Buildings, DBR- N RC Computer Program
No. 35, Marc h 1973.
r
[17] Fothergill, J. W. , e t ai, Development of an Air Move me nt Simulation o
Program , NBS Contract Re port , Int egrated Systems Inc., Roc kvill e ,
Maryland, 1976.

8. Appendix. Heat Load Versus Heat Loss for

Central Heating Systems

When a building is ce ntrally heated , the heating of a To

given space is provided by th e hot supply air to th e sys te m.
The hot air entering the space is usually assumed instantly
(U • A + U . A ) (T - T ) - I . T
mixed with the space air to raise the space temperature. G G w VI A 0

Thus the manner in which the warm space air loses its heat
Q' + H (T
to the cold exterior surfaces present in the space dictates the Ae • He • (Tse - TA) Aw
w sw
heating requirement. In many calculations, however, it is 0'
R -
tacitly assumed that the space air tempe rature is equal to Q
the space surface temperatures and the heating load IS
calculated by a simple formula such as:

FlCURE A-l. A simplified room with all the opaque envelopes having the
same thermal characteristics .
where Vo overall heat transfer coefficient of exterior
surface
room air temperature
outdoor temperature A similar equation ca n be constructed for th e wall portion,
exterior surface areas. again ignoring th e heat storage effect , as follows:

104
 A convec ti o nal way of calcul at ing th e heatin g load by th e
use of Uvalu e is
Ae
+ UWT ' (Tsw - To ) = /. - T (A-2)
Aw

where H w convec tiv e hea t tra nsfe r coeffi c ie nt over th e An exact way of calc ula tin g th e hea lin g load is to say th a I
inle rior sUI{ace of the wall th e heatin g load is th e heat los t by th e roo m a ir to th e
Hwe radiative heat tra nsfer coeffi cie nl of th e inte- surrounding sUlfaces, o r
ri or slllface of th e wall to th e window
UWT th ermal co ndu cta nce be tween th e inte rior sur-
faces of th e wall a nd th e outsid e a mbi e nt
/ sola r radiation inc ide nt upon th e window In order to use th e latter equation th e values of T sw a nd T se
T solar Ira nsm itta nce through th e window mu st be determined by solving eqs (A -I ) and (A-2) s imulta-
Ae window area neo usly. Severa l sampl e ca lc ul a ti o ns a re pe l{ormed as fol-
Aw wall a rea. lows:

The values of UeT a nd UWT can be calcul ated by know in g

th e overall hea l tra nsfe r coeffic ie nt for window Ue and th aI
fo r wall Uw as follows : II is ass umed tha t:

1 1 1 A e = 10, 32, a nd 100 ft 2

VeT Uc He + Hew A w = 900 ft 2
(A-3) T A = 70°F
1 1 1 To = O°F and 40°F
U w = 0.01, O. OS, 0.10 a nd 0. 50 Btu/h ft 2 OF
U WT Uw Hw + Hwe I = 0, 100 a nd 200 Btu/ft2 h
H Re = 0.918 Btu/h ft 2 OF
On th e other ha nd , th e radiative heat transfe r coeffi c ie nts He = H w = 0.542 Btu/h ft 2 OF
Hew a nd H we a re r ela ted to one a nother as follows :
Figure A-2 d e pic ts th e value of R = Q' /Q as affected by
(A-4) the value of U w for the case of I = O. On th e othe r ha nd ,

1.0

TO =0 OF
0.9 TA = 70 OF
"w=900 k '
UG = 1.13 StU / h" OF, k '
CIl
CIl
0.8
0
...J
f--
«
'"
J: 07
0
«
0 "G = 10 k '
...J
0
Z 0.6 "G = 100 k '
f=
«
'"J:
ci 0.5

0.4

0 .3
0.01 0.02 0 .04 0 .06 0 .08 0 .1 0.2 0 .4 0 .5

Uw Btu / hr, ft, OF

FIGURE A-2. Relationship between the heating load Q' and heat loss Q.

105
 TABLE A-I
Aw = 900 ft 2 TA = 70°F Uw = 0

UG/T To AG TSG Tsw Q Q' R

100 0.56/0.78 0 10 43.7 71.3 502 500 1.00
32 44.7 74.1 1613 - 1573 0.98
100 47.8 82.0 5074 - 4650 .92
40 10 59.1 71.6 728 723 .99
32 60.4 75.0 2340 - 2283 .98
100 64.1 84.6 9317 - 6802 .93

1.13/0.87 0 10 16.0 71.0 203 195 .96

32 16.3 73. 1 663 591 .89
100 17.1 78.9 2084 - 1453 .70

40 10 47.0 71.6 658 647 .98

32 47.5 74.9 2112 - 2001 .95
100 48.8 83.9 6607 - 5620 .85

200 0.56/0. 78 0 10 44.4 73.1 1402 - 1390 .99

32 47.0 80.0 4493 - 4388 .98
100 54.2 98.5 - 14074 -13066 .93

40 10 60.0 73.4 - 1630 - 1617 .99

32 62.6 80.7 - 5220 - 5098 .98
100 70.6 101.1 -16317 - 15218 .93

1.13/0.87 0 10 16.2 73.0 - 1203 - 1181 .98

32 17.1 79.4 - 3862 - 3649 .95
100 19.6 96.5 - 12084 - 10198 .84

40 10 47.3 73 .6 - 1658 - 1632 .98

32 48.3 81.1 - 5312 - 5058 .95
100 51.3 101.5 -16607 - 14264 .86

table A-I shows the case of solar heat gains I = 100 and small as long as the heat loss through the wall portion is
200 being included for the calculations. small and the window area is less than 10% of the total
These results show that the error due to the assumption of room surface area. The conclusion is valid even in the case
the same air and exterior surface temperatures is rather where the solar heat gain is included.

106